University of Cambridge Introduces HPC-as-a-Service Offering
The University of Cambridge has implemented a high-performance computing (HPC)-as-a-service
resource to support the scientific and technology research community.
The service runs on the Red Hat OpenStack Platform, and the university has partnered with Red Hat to help bring HPC capabilities to the upstream OpenStack community.
The university's High Performance Computing Service operates two principal high-performance computing clusters: a large CPU
cluster known as Darwin, and an energy-efficient GPU cluster known as
Wilkes. In an effort to extend its high performance and research
computing services to the broader scientific and technology research
community, the university has established a new HPC-as-a-service
offering.
Cambridge's HPC-as-a-service runs on the OpenStack
open source software for creating public and private cloud-computing
services. During the proof-of-concept phase of the project, the
university implemented OpenStack on a community-supported Linux
distribution. Ultimately they determined that they would need "a more
reliable, integrated and supported OpenStack platform for production
deployment," according to a news release. At that point, the university
turned to the Red Hat OpenStack Platform, which is "a highly scalable,
production-ready Infrastructure-as-a-Service (IaaS) solution,"
according to the company.
As part of the collaboration between
the University of Cambridge and Red Hat, the organizations plan to
contribute HPC capabilities to the upstream OpenStack community, which
will help "to advance the open source cloud platform for scientific
research use cases," according to Red Hat.
About the Author
Leila Meyer is a technology writer based in British Columbia. She can be reached at [email protected].